One way that spatial locations of sound sources may be resolved is by a listener perceiving an interaural level difference (“ILD”) of a sound at each of the two ears of the listener. For example, if the listener perceives that a sound has a relatively high level (i.e., is relatively loud) at his or her left ear as compared to having a relatively low level (i.e., being relatively quiet) at his or her right ear, the listener may determine, based on the ILD between the sound at each ear, that the spatial location of the sound source is to the left of the listener. The relative magnitude of the ILD may further indicate to the listener whether the sound source is located slightly to the left of center (in the case of a relatively small ILD) or further to the left (in the case of a larger ILD). In this way, listeners may use ILD cues along with other types of spatial cues (e.g., interaural time difference (“ITD”) cues, etc.) to localize various sound sources in the world around them, as well as to segregate and/or distinguish the sound sources from noise and/or from other sound sources.
Unfortunately, many binaural hearing systems (e.g., cochlear implant systems, hearing aid systems, earphone systems, mixed hearing systems, etc.) are not configured to preserve ILD cues in representations of sound provided to users relying on the binaural hearing systems, and, as a result, it may be difficult for the users to localize sound sources around themselves or to segregate and/or distinguish particular sound sources from other sound sources or from noise in the environment surrounding the users. Even binaural hearing systems that attempt to encode ILD cues into representations of sound provided to users have been of limited use in enabling the users to successfully and easily localize the sound sources around them. For example, some binaural hearing systems have attempted to detect, estimate, and/or compute ILD and/or ITD spatial cues, and then to convert and/or reproduce the spatial cues to present them as ILD cues to the user. Unfortunately, the detection, estimation, conversion, and reproduction of ILD and/or ITD spatial cues tend to be difficult, processing-intensive, and error-prone. For example, noise, distortion, signal processing errors and artifacts, etc., all may be difficult to control and account for in techniques for detecting, estimating, converting, and/or reproducing these spatial cues. As a result, when imperfect spatial cues are presented to users of binaural hearing systems due to these difficulties, the users may inaccurately localize sound sources or be disoriented, confused, and/or misled by conflicting or erroneous spatial cues. For example, a user may perceive that a sound source is moving around when the sound source is actually stationary.
Moreover, independent signal processing at each ear (e.g., various types of gain processing such as automatic gain control, noise cancelation, wind cancelation, reverberation cancelation, impulse cancelation, and the like, performed by respective sound processors at each ear) may deteriorate spatial cues even if the spatial cues are detected, estimated, converted, and/or reproduced without errors or artifacts. For example, a sound coming from the left of the user may be detected to have a relatively high level at the left ear and a relatively low level at the right ear, but that level difference may deteriorate as various stages of gain processing at each ear independently process the signal (e.g., including by adjusting the signal level) prior to presenting a representation of the sound to the user at each ear.